Planning a Radioactive Injection and an X-ray Beam Together One Treatment Plan for Two Radiation Sources

This application covers a method for writing a single radiotherapy plan for a patient who will receive radiation from two directions at once: a radioactive drug injected into the body and a beam of radiation aimed at the tumour from outside. Rather than planning each separately and adding them up, the method adjusts both doses together until the combined dose meets the prescription. It was filed by Reflexion Medical, Inc., names Peter Demetri Olcott, Michael Kirk Owens and Debashish Pal as inventors, and is a divisional of Australian application 2021212659.

The Problem

The specification sets out the strengths and weaknesses of each kind of radiation. External beam radiotherapy can deliver a very even dose to a solid tumour that shows up on scans, but many cancers are not visible on pre-treatment images, so a beam cannot cure widely spread or microscopic disease. An internal source, such as an injected radiopharmaceutical, travels through the body and can reach scattered or microscopic deposits, but it also collects in healthy tissue, which can be toxic, and it tends to pile up in the middle of a large tumour and thin out toward the edge.

Using both therefore makes sense, and the specification describes the obvious way to combine them: plan the injected dose and the beam dose separately, then scale each by a factor and add them. The problem is what that does to the dose map. In the specification’s simulation of that approach, the injected dose gave more than about 25 Gy to less than 10 per cent of the planning target volume, concentrated at its centre, while the beam alone gave more than about 50 Gy to the whole volume. Scaling and summing the two keeps the lumpy pattern left by the injection, so the combined dose is no more even than the injection alone.

What This Invention Does

The title and claim 1 agree here. Claim 1 is a method for generating a joint internal and external radiotherapy plan. It calculates the dose that can be delivered with the internal source, calculates the dose that can be delivered with the external source, adjusts either or both to reach a cumulative dose that meets the prescribed requirements for a patient target region, and then generates a plan that specifies the dose from each source such that the two add up to that cumulative dose.

Claim 1 on its own is broad. The way the specification gets a better dose map is described in the dependent claims. Claim 26 adjusts the two doses by iterating through different values of the internal dose together with different values of the external dose until one or more dose constraints are met. Claim 8 expresses the internal dose as a matrix R that maps the quantity of drug given to the dose in each patient region, and claim 16 expresses the beam dose as a matrix A that maps the beam fluence to the dose in each region. Claim 28 then solves for the drug quantity and the beam fluence together so that the cost functions are met for the sum of the two. Claims 29 to 34 build those cost functions from weighted terms, with the highest weight reserved for the top priority objective, and claim 31 includes a cost for toxicity to healthy tissue.

In the specification’s joint optimisation example, the beam is deliberately lowered toward the centre of the target, giving what it calls a “donut” shaped beam dose, which fills in around the injected dose rather than overlapping its hot spot. The combined distribution is described as more homogeneous, with a rapid fall-off outside the target, and it reports the dose coverage figures for the target volume in its dose-volume histograms.

Key Features

• One plan for two sources. The plan states a dose for the injected source and a dose for the beam source, chosen so that they sum to the cumulative dose.
• Image driven internal dose. Claims 2 to 6 calculate the internal dose from functional image data, including positron emission tomography (PET), and claim 6 lists 27 imaging or therapy agents, among them F-18, Ga-68, Cu-64, Lu-177, Y-90, I-131, Ac-225 and Ra-223.
• Same targeting molecule for scan and treatment. Claim 9 calculates the dose from images taken with a diagnostic compound that has the same targeting backbone as the therapeutic one. The specification’s example pairs Ga-68 DOTA-TATE for imaging with Lu-177 DOTA-TATE for treatment.
• Matrix model of each dose. Internal dose is written as R times q and beam dose as A times x, so a standard optimiser can solve for the drug quantity q and the beam fluence x together.
• Weighted, ranked goals. Cost functions on dose, quantity and toxicity are combined with weights that reflect priority. The specification gives clinical objective examples such as keeping the maximum kidney dose below 40 Gy.
• Biologically equivalent dose. Claim 35 allows the cumulative dose to be a biologically equivalent dose (BED), which accounts for how differently the two radiation types affect tissue.
• Beam guided by PET. Claims 23 to 25 cover a radiotherapy system with PET detectors that applies radiation from positron emission data, using a tracer with the same targeting backbone as the injected drug.

Who Is Behind It

The applicant is Reflexion Medical, a radiation therapy company. It makes a linear accelerator system, the X1, which carries PET detectors, and which supports a treatment mode called biology-guided radiotherapy, in which radiation beamlets are steered in real time by PET signals coming from the tumour. A Physics World report on the first-in-human study of that mode describes plans that met clinical criteria for all nine patients in one cohort, and says that researchers plan to evaluate other tracers, including PSMA. The specification refers to the same technique as biologically-guided radiation therapy (BGRT), which is why the claims bring PET detectors into the picture.

On priority, the specification says it claims priority to United States provisional application 62/966,997, filed 28 January 2020. The page headers carry the numbers WO 2021/154746 and PCT/US2021/015119, and the title page records this as a divisional of Australian application 2021212659. The priority country is therefore the United States. The title page does not give the applicant’s address, so the applicant country in the table is inferred from the company and the priority filing, and is not stated in the document.

The inventors have no web profile that could be verified from the document alone, and the article does not describe them further.

Why It Matters

Radioligand therapies are growing quickly, and many patients who receive them also have lesions that a beam could treat. Today the two usually sit in different departments, with a nuclear medicine team deciding the drug dose and a radiation oncology team planning the beam. This claim describes a planning step that puts both in the same calculation, with a shared set of constraints for organs at risk such as the kidneys.

The practical argument is homogeneity. A tumour with an under-dosed rim can regrow from its edge, and the specification’s simulation shows how a beam that is told about the injected dose can be shaped to cover the gaps. The method also supports adapting between sessions, since the specification describes using images from a previous session to adjust the dose given in a later one.

What the document does not supply is clinical outcome data. The numbers it gives come from planning simulations, not from patients, and the claim says nothing about which cancers or drugs the method is for beyond the list of radionuclides.

Related Concepts

• Radiation therapy – the broad field, including the external beam half of the claimed method.
• Peptide receptor radionuclide therapy – an injected treatment of the kind the internal source stands for.
• Positron emission tomography – the imaging method that supplies the functional data in the dependent claims.
• Lutetium-177 – the beta emitting radionuclide the specification uses as its main example.
• Dotatate – the targeting molecule the specification pairs with Ga-68 for imaging and Lu-177 for treatment.
• Biologically effective dose – the dose measure that claim 35 allows for the cumulative dose.

---

AU 2026202282 was published in the Australian Official Journal of Patents on 23 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.